Project Summary While the risk of Alzheimer?s disease (AD) increases with advancing age, approximately 5% of AD patients develop symptoms before age 65 (~280,000 Americans). The vast majority (90%-95%) of EOAD patients do not have a known mutation in APP or PSEN1/2, and only ~50% are APOE4 carriers. Unlike late-onset AD (LOAD), 30-64% of EOAD have non-amnestic presentations, leading to missed or delayed diagnosis. Despite being highly motivated and having few comorbidities, EOAD patients are commonly excluded from large scale observational biomarker studies (e.g. ADNI and DIAN) and therapeutic trials due to their young age, non- amnestic deficits, or absence of known pathogenic mutations. Furthermore, studies suggest high heritability in EOAD in the absence of known mutations or APOE4, signifying that this population may be enriched for novel genetic risk factors. Emerging biomarkers of amyloid and tau have not been systematically characterized in this population. Clinical and neuroimaging measures employed in LOAD may be insensitive to baseline deficits and disease progression in EOAD, which predominantly involve non-memory cognitive domains and posterior cortical neurodegeneration. To fill this gap in AD research, we plan to recruit and longitudinally follow 400 amyloid PET- positive EOAD subjects meeting NIA-AA criteria for MCI due to AD or probable AD dementia (including primary amnestic, dysexecutive, language and visuospatial presentations) and 100 age-matched controls. Participants in the Longitudinal Early-onset Alzheimer?s Disease Study (LEADS) will undergo clinical assessments, psychometric testing, MRI, amyloid ([18F]Florbetaben) and tau ([18F]AV1451) PET, CSF and blood draw for collection of DNA, RNA, plasma, serum and peripheral blood mononuclear cells (PBMC). Patients will be assessed at three time points ? baseline (both EOAD and controls), 12 months (EOAD all measures; controls ? clinical and cognitive measures only) and 24 months (EOAD, all measures except PET). Methods will be harmonized with ADNI and DIAN. We will comprehensively characterize cognitive, imaging and biofluid changes over time in EOAD, and compare to a matched sample of LOAD participants identified in ADNI. We will employ machine learning algorithms to develop sensitive clinical and imaging measures of EOAD progression. An exploratory aim will apply next generation sequencing to assess for novel genetic risk factors for disease. The study will also establish a network of EOAD research sites and set the stage for the launch of clinical trials in this population.